Natural language processing (NLP) is a subfield of semantics, computer science, software engineering, and artificial intelligence interested about the cooperation among PCs and human language, specifically how to program PCs to measure and dissect a lot of natural language data.
Some applications of NLP :
text translation (DeepL for example)
automatic summary of content
opinion / sentiment analysis
next word prediction on smartphone
extracting named entities from text
So the main aim of NLP i build systems that can make sense of text and perform tasks already seen like translation, grammar checking, or topic classification. The most popular example for using NLP is Google Assistant it help to reply to some tasks like : ‘’Hi! Where is the nearest gas station ?’’ into numbers.
Another famous example for NLP using is chatboots, you can find them in many website where you get a service like Ebay and Alibaba and Payoneer, when you need more information and you try to contact customer service by using live chat, first a boot talk with you trying to understand your task and do the best to redirect you to help page if you asked about an popular task without passing to an human agent to assist you.
Why NLP is growth everyday speedily?
The secret of this big evolution of NLP is that data is everywhere, as branch of Artificial Intelligence data is so important to build model train it and test, so no worry if we are work with human language, billion tons of data are available to use, there is unlimited number of words and languages
What is difference between NLP and Machine Learning?
Artificial Intelligence is a umbrella term for automatic systems or ‘’Machines’’ which are trying to be dynamic and understand by learning from large datasets and try to solve news problems: prediction, classification, detection, translate,,,,
We can summarize the NLP process with this following diagram :
Why NLP use Machine Learning?
As we seen before in the diagram above, the process of NLP look like an transfer function with input and output, so we can use ML algorithm to process NLP tasks by using data and control the output.
NLP can be classified into two classes: NLU for Natural Language Understanding, NLG for Natural Language Generation.
Note that linguistic is the science of language, which is divided into Phonology which is text sound, Morphology which is text formation, Syntax sense, semantic syntax and pragmatics which mean understanding.